|Publication number||US5187598 A|
|Application number||US 07/749,752|
|Publication date||Feb 16, 1993|
|Filing date||Aug 26, 1991|
|Priority date||Aug 26, 1991|
|Publication number||07749752, 749752, US 5187598 A, US 5187598A, US-A-5187598, US5187598 A, US5187598A|
|Inventors||Thomas J. Posluszny, Dennis W. Davis|
|Original Assignee||United Technologies Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (13), Non-Patent Citations (8), Referenced by (11), Classifications (16), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Technical Field
The field of the invention is that of optical signal processing, in particular, the identification of signals amidst background noise.
2. Background Art
The problem of identifying a particular pulse pattern and separating it from random background noise or from other nonrandom signals is an old one in the art. Electrical methods, such as the Watkins-Johnson pulse interval processor, have limited throughput and are thus at a disadvantage in the real-time processing that is usually required. A number of attempts have been made in the art to use optical processing for this problem, but none of them have become accepted.
The invention relates to an optical processing system for deinterleaving a pulse train from background clutter. The signal data is displayed on a two-dimensional spatial light modulator (SLM) at a frequency approximately equal to the expected frequency of the pulse being looked for. The sought-after repetitive pulse signal will be displayed as a line on the SLM. Pattern recognition of this line among the background is accomplished by taking the Fourier transform of the amplitude distribution of the radiation emitted from the SLM; passing that through a holographic optical element that performs a rectangular to polar coordinate conversion; taking the Fourier transform of the output from the holographic element and passing that through a line spatial filter oriented at a right angle to the linear output of the holographic element; inverse Fourier transforming this result; and then focusing it onto a linear detector array.
Other features and advantages will be apparent from the specification and claims and from the accompanying drawings which illustrate an embodiment of the invention.
FIG. 1 illustrates schematically an embodiment of the invention.
FIGS. 2a through 2f illustrate a series of representations of the energy distribution in the optical beam passing through the system.
FIG. 3 illustrates in part pictorial, part schematic form an embodiment of the invention.
Referring now to FIG. 1, an incoming pulse train of signals 210 is shown as entering box 250 on line 205. Box 250 represents an SLM and associated electronics for addressing and controlling the pixels, etc. The data is displayed in line sequential raster scan format on the device. If the data display rate or effective "sweep frequency" across the SLM is the same as the repetition frequency of the incoming data, the displayed data or modulated pixels will be displayed on the face of the SLM as a vertical line, as shown in FIG. 1. If the sweep rate is somewhat different from that of the incoming data, the line will be at an angle. There will, of course, be background noise superimposed on the SLM, whether it is random electrical noise or other pulses from all the background signals that are inevitably present.
In the illustrative embodiment, a pixel in the SLM becomes transparent when there is a pulse at the corresponding time so that the energy distribution is a series of dots of light. The opposite approach can also be taken in which data is expressed as a dark pixel against a light background. In the example of FIG. 2a, the data is shown as having a significant angle and being offset by an amount that relates to the phase and frequency differences between the incoming data and the sweep cycle of the SLM. The operation of the SLM will be referred to as presenting an array of pixels, since the pixels are "presented" to an optical beam, for ex. a laser beam that will be modulated by transmission through the SLM to emerge as a modulated laser beam having an amplitude modulated wavefront that carries a representation of the data pattern formed by the pixels on the SLM.
The phrase "carry a representation" is used here as a general term meaning that the amplitude and/or phase of the beam has been modified so that there is a connection between the data and the amplitude and/or phase. The relationship need not be linear and in general will not be.
An optical wavefront having an intensity modulated according to this pattern must be processed in order to move the desired signal from whatever random position it initially assumes to a standard position that can be accepted by the remainder of the system. In a next processing stage, the modulated beam undergoes a coordinate transformation from Cartesian to polar representation that places it in a standard form for recognition, namely a laterally displaced vertical line. For convenience in the terminology, the rows of pixels in SLM define the horizontal, with the vertical being at a right angle. SLM can be rotated at any angle with respect to the usual vertical and horizontal directions.
This laterally displaced vertical line is then correlated with a horizontal slit mask (which constitutes a matched filter for this input). The correlation energy distribution can then be focused into a linear set of detectors in order to provide a manageable task for conventional electronic signal processing.
The details of the aforementioned optical processing are explained with reference to FIG. 3, a schematic of the sequential processing elements, and FIG. 2, a corresponding sequence of image representations of the processed wavefront. Referring now specifically to FIG. 3, there is illustrated a system in which a collimated laser beam 1 from laser 100 is introduced from the left and passes through a transmissive spatial light modulator (SLM) 2, illustratively an array of magneto-optic modulators such as the 128×128 pixel array from SEMETEX Corp. This SLM imparts an intensity modulation across the beam representing the pulse train data image as depicted in FIG. 2a. The Fourier transform of this image is provided at the front face of an optically-addressed spatial light modulator (OASLM) 4 by lens 3. The magnitude of this transform is replicated on the back face of the OASLM as a reflectivity map corresponding to FIG. 2b which depicts a line oriented perpendicular to the input line. This is the case because for any given laterally displaced input object, the amplitude of its Fourier transform remains invariant whereas the Fourier transform phase is altered. Hence, in the instance of a line object, its Fourier transform amplitude profile (which happens to be a line orthogonal to the input line) is constrained to pass through the origin of the Fourier plane independent of shifts in the position of the input line object. By virtue of reflection, this transform magnitude is impressed upon a collimated beam from read laser 8 which is introduced to the back face of the OASLM by via beamsplitter 7. The reflected beam from laser 8 travels to the right carrying the transform magnitude. The beam may, of course, be from laser 100.
Elements 5,6,9,10, and 11, referred to as polar coordinate conversion means for convenience, constitute the set of optics which performs a Cartesian to log-polar (the transform to log-polar is conformal and therefore easier to implement than transformation to a pure polar output and it does not impact the function of the overall system for the purpose of recognizing linear features) coordinate transformation of the intensity appearing on the back face of the OASLM (i.e., a representation of the Fourier transform magnitude of the input image). In the log-polar plane a short vertical streak of energy will occur at a position along the abscissa corresponding to the angular orientation of the line in the Cartesian plane.
The tandem filter approach to this coordinate transform is detailed in Hossack, et al., "Coordinate Transformations with Multiple Computer-generated Optical Elements", Journal of Modern Optics, Vol. 34, No. 9, 1987, pg. 1235-1250. Two variations to the Hossack approach are used herein. The first of these is the use of two lenses, 9 and 11, which implement quadratic phase terms in the transformation that would otherwise require some of the resolution available in the holographic optical elements, 6 and 10. With the use of this approach, a hologram of a given spatial resolution will produce a better transform. The second point of departure is related to the fact that elements 5 and 6 are placed at the plane of the OASLM output so as to avoid the need for additional relay optics. Since the read laser traverses both of these elements twice, the single pass amplitude mask is replaced with its square root and the phase mask is reduced by a factor of 2.
The density distribution in amplitude mask 5 is
A=(xi 2 +yi 2)-1/4
The phase shift distribution of mask 6 is ##EQU1## and the phase shift distribution of mask 10 is ##EQU2##
The log-polar map of the Fourier transform magnitude of the input image is present at plane 12 and corresponds to FIG. 2c. It is at this plane that all linear features present in the input image will appear as vertical lines at various lateral displacements. These displacements correspond to the angular orientations of input line features and hence the various data pulse repetition rates represented in the input image. Those skilled in the art will readily appreciate that a detector could be installed at plane 12 to measure the radiation; e.g. by having a two-dimensional photodetector array with lines of vertical pixels wired to sum their outputs. Such a detector would also detect the miscellaneous pulses that will inevitably be present and an amplitude discriminator will be used to differentiate between vertical lines having a number of energized pixels above and below some threshold.
Correlation of the distribution at plane 12 with a matched filter 14 for vertical lines (a horizontal slit shown in FIG. 2d) is achieved with Fourier transform lenses 13 and 15 and provides an output from lens 15 depicted in FIG. 2e. When this distribution is collapsed into one dimension by cylindrical lens 16, a correlation peak in energy occurs at the horizontal position of the aforementioned vertical line at the plane of detector array 17 as shown in FIG. 2f. Elements 13-17 will be referred to as a linear correlation unit for convenience in terminology. The deviation of the pulse repetition frequencies in the input data from a nominal frequency is indicated by the lateral position of this peak.
The input SLM can be a high speed, electronically addressed, binary transmissive device such as a 128×128 pixel magneto-optic SLM available from SEMETEX Corporation of Torrance, California. This device provides a nominal frame rate of 500 Hertz with burst mode operation a 2000 frames/second and a contrast ratio of 104:1. The OASLM can be equally as fast or faster (10's of microseconds frame time) as is disclosed in Maserjian, et al. "Optically addressed spatial light modulators by MBE-grown nipi MQW structures", Applied Optics, Vol. 28, No. 22, Nov. 15, 1989, pp. 4801-4807.
It should be understood that the invention is not limited to the particular embodiments shown and described herein, but that various changes and modifications may be made without departing from the spirit and scope of this novel concept as defined by the following claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3509565 *||Jun 19, 1961||Apr 28, 1970||Raymond M Wilmotte||Method and apparatus for optically processing information|
|US3903407 *||Dec 11, 1963||Sep 2, 1975||Us Navy||Method for correlating frequency-modulated signals|
|US4071907 *||Oct 12, 1976||Jan 31, 1978||The United States Of America As Represented By The Secretary Of The Navy||Radar signal processor utilizing a multi-channel optical correlator|
|US4073010 *||Jul 23, 1976||Feb 7, 1978||The United States Of America As Represented By The Secretary Of The Navy||Correlation methods and apparatus utilizing mellin transforms|
|US4084255 *||Nov 2, 1976||Apr 11, 1978||The United States Of America As Represented By The Secretary Of The Navy||Positional, rotational and scale invariant optical correlation method and apparatus|
|US4099249 *||Feb 22, 1977||Jul 4, 1978||The United States Of America As Represented By The Secretary Of The Navy||Doppler processing method and apparatus|
|US4389093 *||Mar 23, 1981||Jun 21, 1983||Probe Systems, Inc.||Method and apparatus for coherent detection in optical processors|
|US4487476 *||Apr 28, 1981||Dec 11, 1984||The United States Of America As Represented By The Secretary Of The Air Force||Method of multivariant intraclass pattern recognition|
|US4628473 *||Jul 27, 1984||Dec 9, 1986||Cooper Lasersonics, Inc.||System for autocorrelating optical radiation signals|
|US4636718 *||Jul 20, 1984||Jan 13, 1987||Sperry Corporation||Acousto-optical spectrum analyzer with expanded frequency resolution|
|US4712059 *||Nov 12, 1985||Dec 8, 1987||Sperry Corporation||Broadband optical processor for determination of frequency and time of arrival of multiple signals|
|US4826285 *||Sep 24, 1987||May 2, 1989||Horner Joseph L||Method of enhancing the signal to noise ratio of an image recognition correlator|
|US5005946 *||Apr 6, 1989||Apr 9, 1991||Grumman Aerospace Corporation||Multi-channel filter system|
|1||"Real-Time Deformation Invariant Optical Pattern Recognition Using Coordinate Transformations", Casasent et al. Applied Optics, vol. 26 No. 5, Mar. 1987. pp. 938-942.|
|2||Hossack et al., "Coordinate Transformations with Multiple Computer-Generated Optical Elements", Journal of Modern Optics, vol. 34, No. 9 1987, pp. 1235-1250.|
|3||*||Hossack et al., Coordinate Transformations with Multiple Computer Generated Optical Elements , Journal of Modern Optics, vol. 34, No. 9 1987, pp. 1235 1250.|
|4||Maserjian, et al., "Optically Addressed Spatial Light Modulators by MBE-Grown Nipi MQW Structures", Applied Optics, vol. 28, No. 22, Nov. 15, 1989, pp. 4801-4807.|
|5||*||Maserjian, et al., Optically Addressed Spatial Light Modulators by MBE Grown Nipi MQW Structures , Applied Optics, vol. 28, No. 22, Nov. 15, 1989, pp. 4801 4807.|
|6||*||O Meara et al., Optical Phase Conjugation pp. 559, (1983).|
|7||O'Meara et al., Optical Phase Conjugation pp. 559, (1983).|
|8||*||Real Time Deformation Invariant Optical Pattern Recognition Using Coordinate Transformations , Casasent et al. Applied Optics, vol. 26 No. 5, Mar. 1987. pp. 938 942.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US5289299 *||Jan 15, 1993||Feb 22, 1994||Bell Communications Research, Inc.||Holographic code division multiple access|
|US5528389 *||Nov 14, 1994||Jun 18, 1996||At&T Corp.||Optical holographic system for parallel to serial and serial to parallel conversion of optical data|
|US5682262 *||Dec 13, 1995||Oct 28, 1997||Massachusetts Institute Of Technology||Method and device for generating spatially and temporally shaped optical waveforms|
|US5736958 *||Apr 15, 1997||Apr 7, 1998||Essex Corporation||Image synthesis using time sequential holography|
|US5914802 *||Jul 18, 1997||Jun 22, 1999||Northrop Grumman Corporation||Combined spatial light modulator and phase mask for holographic storage system|
|US6266452 *||Mar 18, 1999||Jul 24, 2001||Nec Research Institute, Inc.||Image registration method|
|US7791560||Jan 29, 2008||Sep 7, 2010||Mark Anthony Gleeson Smith||Image transfer apparatus|
|US8284469 *||Oct 9, 2012||Thomson Licensing||Apparatus for reading from and/or writing to holographic storage media|
|US20080247013 *||Jan 29, 2008||Oct 9, 2008||F. Poszat Hu, Llc||Image transfer apparatus|
|US20080259420 *||Mar 20, 2008||Oct 23, 2008||Wolfgang Hossfeld||Apparatus for reading from and/or writing to holographic storage media|
|WO2008094948A1 *||Jan 29, 2008||Aug 7, 2008||F. Poszat Hu, Llc||Image transfer apparatus|
|U.S. Classification||359/29, 359/563, 359/559, 708/816|
|International Classification||G06E3/00, G02B5/32, G01R23/17, G02B27/46|
|Cooperative Classification||G02B5/32, G01R23/17, G02B27/46, G06E3/005|
|European Classification||G02B5/32, G02B27/46, G01R23/17, G06E3/00A2|
|Aug 26, 1991||AS||Assignment|
Owner name: UNITED TECHNOLOGIES CORPORATION, CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNORS:POSLUSZNY, THOMAS J.;DAVIS, DENNIS W.;REEL/FRAME:005830/0445;SIGNING DATES FROM 19910805 TO 19910810
|Sep 24, 1996||REMI||Maintenance fee reminder mailed|
|Feb 16, 1997||LAPS||Lapse for failure to pay maintenance fees|
|Apr 29, 1997||FP||Expired due to failure to pay maintenance fee|
Effective date: 19970219